| Chapter I Three-dimensional Contrast Enhanced Ultrasound in the Diagnostic Strategy of Tumor Angiogenesis in Breast CancerPart I Three-dimensional Contrast Enhanced Ultrasound in the Assessment of Breast Tumor AngiogenesisObjective:To differentiate the perfusion characteristics and tumor vascular characteristics between benign and malignant breast tumors by three-dimensional contrast enhanced ultrasound (3D-CEUS), and to explore the clinical diagnostic value of3D-CEUS, which may be a new radiologic modality in the evaluation of breast tumor angiogenesis. Methods:183patients with breast tumor from May2011to February2012were included. One male,182female, aged20to76years, with an average of50.25years old. Esaote Mylab90(Esaote, Genoa, Italy) was performed with SonoVue (SonoVue, Bracco, Italy). Combined with multiple display modes of three-dimensional ultrasound, the five continuous images obtained by3D-CEUS were referred to as S1, S2, S3, S4, S5, respectively. The strongest perfusion stage and the beginning to washout stage were recorded. Then the characteristics of tumor vascularity and the intratumoral perfusion in the strongest perfusion stage was focused, including the presence, the number, distribution, course and dilated degree of peripheral vessels; the presence and the coarse degree of rim perfusion; the presence of penetrating vessels; the presence of internal perfusion as well as internal perfusion type; the presence of central vessels, as well as the dilated degree of central vessels; the presence of the internal perfusion defects and perfusion defect ratio; the presence of residual perfusion in S5stage. Results:183cases of breast tumors,35cases were benign,148were malignant. The number (P=0.000), distribution (P=0.000), courses (P=0.000), and dilated degree of peripheral vessels (P-0.000), presence of rim perfusion (P=0.018), coarse degree of rim perfusion (P=0.000), presence of penetrating vessels (P=0.006), internal perfusion type (P=0.000), dilated degree of central vessels (P=0.008), perfusion strongest stage (P=0.000), and beginning to washout stage (P=0.000) were all statistically significant different between benign and malignant breast tumors. The presence of peripheral vessels, central vessel, internal perfusion, internal perfusion defects, residual perfusion in S5, and the perfusion defect ratio were not statistically significant different between benign and malignant (P>0.05). The sensitivity, specificity, and accuracy in the number of peripheral vessels were79.7%,72.7%, and78.4%, respectively.The sensitivity, specificity, and accuracy of peripheral radial vessels in the diagnosis of breast cancer were82.6%,100.0%, and86.0%, respectively. The sensitivity, specificity, and accuracy of peripheral vessels with;, tortuous courses were66.7%,97.0%, and72.5%, respectively. As for dilated peripheral vessels or caliber fluctuations, the sensitivity, specificity, and accuracy were81.2%,60.6%, and77.2%, respectively. The sensitivity, specificity, and accuracy of coarse degree of rim perfusion were90.2%,70.4%, and85.3%, respectively. The sensitivity, specificity, and accuracy of penetrating vessels were35.1%,88.6%, and45.4%, respectively. The sensitivity, specificity, and accuracy of internal perfusion type were77.8%,57.1%, and73.5%, respectively. The sensitivity, specificity, and accuracy of dilated degree of central vessels were38.0%,88.9%, and48.8%, respectively. The sensitivity, specificity, and accuracy of perfusion strongest stage in S1were75.7%,60.0%, and72.7%, respectively. The sensitivity, specificity, and accuracy of beginning to washout stage in S2were70.3%,80.0%, and72.1%, respectively. The correlation between pathology with distribution(r=0.692, P=0.000), number (r=0.450, P=0.000), course (r=0.504, P=0.000), dilated degree (r=0.372, P=0.000) of peripheral vessels was positive; as well as coarse degree of rim perfusion(r=0.606, P=0.000); the presence of penetrating vessels (r=0.202, P=0.006), the internal perfusion type (r=0.310, P=0.000), the dilated degree of central vessels (r=0.235, P=0.008). The correlation between pathology with the presence of rim perfusion, strongest perfusion stage and the beginning to washout stage was negative. Conclusion:3D-CEUS with second-generation ultrasound contrast agent(SonoVue), can objectively evaluate the characteristics of breast tumor angiogenesis in order to improve the value of ultrasound in differentiating benign and malignant breast tumors. Part Ⅱ Correlation between Prognostic Factors with Three-dimensional Contrast Enhanced Ultrasound in Breast TumorsObjective:To explore the correlation between perfusion characteristics of breast tumors by three-dimensional contrast enhanced ultrasound (3D-CEUS) with tumor size, lymph nodes, pathological type, histological grade, estrogen receptor (ER), progesterone receptor (PR), CerbB-2(also known as neu or Her-2gene), Ki-67, microvessel density (MVD), vascular endothelial growth factor (VEGF), matrix metalloproteinases (MMPs) in order to assess the possibility of indirectly reflection biological behavior of breast cancer by3D-CEUS, and to provide reliable information for predicting the prognosis of breast cancer as well as developing a reasonable individualized treatment scheme. Methods:Esaote Mylab90(Esaote, Genoa, Italy) was performed with SonoVue (SonoVue, Bracco, Italy).183patients with breast tumors, divided into two groups by conventional ultrasound maximal diameter measurement, respectively, along with MVD, VEGF, MMP-2and MMP-9staining. As for breast cancer patients, ER, PR, CerbB-2, and Ki-67immunohistochemical staining were additionally performed. Results:MVD (P=0.000), VEGF (P=0.000), MMP-2(P=0.000) and MMP-9(P=0.000) were statistically significant different between benign and malignant breast tumors. The best cutoff value of MVD was7001.67. The sensitivity, specificity, and accuracy in the diagnosis of breast cancer were87.8%,100.0%, and90.2%, respectively. The area under the curve of MVD was0.939(95%CI:0.906～0.972). The correlation between MVD and pathology was significant (r=0.76,P=0.000). The sensitivity, specificity, and accuracy of VEGF in the diagnosis of breast cancer were86.5%,74.3%, and84.2%, respectively. The area under the curve of VEGF was0.804(95%CI:0.71～0.89). The correlation between VEGF and pathology was significant(r=0.55, P=0.000). The sensitivity, specificity, and accuracy of MMP-2in the diagnosis of breast cancer were79.7%,65.7%, and77.0%, respectively. The area under the curve of MMP-2was0.727(95%CI:0.628～0.827). The correlation between MMP-2and pathology was low(r=0.39, P=0.000). The sensitivity, specificity, and accuracy of MMP-9in the diagnosis of breast cancer were90.5%,48.6%, and82.5%, respectively. The area under the curve of MMP-9was0.696(95%CI:0.586-0.805). The correlation between MMP-9and pathology was low(r=0.41, P=0.000).The tumor size was statistically significant different between benign and malignant breast tumors. The number of peripheral vessels (P=0.000), the radial distribution (P=0.000), tortuous courses (P=0.001), dilated degree (P=0.021), the coarse degree of rim perfusion (P=0.001), the presence of penetrating vessels (P=0.003), internal perfusion type(P=0.001), the dilated degree of central vessels(P=0.000), the perfusion defect ratio (P=0.002), and beginning to washout stage (P=0.001) were all statistically significant different between the two different tumor size groups. Many3D-CEUS parameters were statistically significant different between invasive ductal cancer (IDC) and fibroadenoma, while only some of the3D-CEUS parameters were statistically significant different between IDC and non-IDC. Only some of the3D-CEUS parameters were statistically significant different between different lymph node status, histological grade, ER, PR, CerbB-2and Ki-67expression. The number (P-0.000), the radial distribution (P=0.000), tortuous courses (P=0.000), dilated degree (P=0.000) of peripheral vessels, and the coarse degree of rim perfusion (P=0.000) were all statistically significant different between different MVD, VEGF, MMP-2, and MMP-9expression. Conclusion:The performance of3D-CEUS, particularly in the peripheral vessels characteristics, closely related to tumor size, pathological type, and the expression of MVD, VEGF MMP-2, and MMP-9, can indirectly reflected biological behavior.3D-CEUS can be a new non-invasive and in vivo method to assess breast tumor angiogenesis and breast cancer prognosis. Part Ⅲ Comparative Study of Three-dimensional Contrast Enhanced Ultrasound and Dynamic Contrast-enhanced Magnetic Resonance Imaging in the Diagnosis of Breast Tumor AngiogenesisObjective:To explore the clinical feasibility and diagnostic value of three-dimensional contrast enhanced ultrasound (3D-CEUS) score in the evaluation of breast tumor angiogenesis, compared with dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) score. Methods:183patients with breast tumor from May2011to February2012were included. One male,182female, aged20to76years, with an average of50.25years old. Esaote Mylab90(Esaote, Genoa, Italy) was performed with SonoVue (SonoVue, Bracco, Italy). The establishment of3D-CEUS score system was based on the characteristics as follows:the presence, course and dilated degree of peripheral vessels, the presence of penetrating vessels, the presence and the coarse degree of rim perfusion, and the presence of central vessels, as well as the dilated degree of central vessels, and tumor vascular characteristics between benign and malignant breast tumors, which were recorded as0,1,2points, respectively. DCE-MRI was-performed on a dedicated breast magnetic resonance imaging (DBMRI) system (Aurora Dedicated Breast MRI Systems, USA) with Gd-DTPA (Magnevist, Germany). According to the Fischer score, the establishment of DCE-MRI score system was appropriately modified as follows:enhancement type, time-signal intensity curve (TIC), early phase enhancement rate (EPER), peripheral vessels, and penetrating vessels. Results:3D-CEUS score in malignant breast tumors was statistically significant different higher than benign tumors (P=0.000). The sensitivity, specificity, and accuracy of3D-CEUS score with4points in the diagnosis of breast cancer were85.1%,94.3%, and86.9%, respectively. The correlation between3D-CEUS score and pathology was significant (r=0.68, P=0.000). The correlation between3D-CEUS score and MVD was significant (r=0.50, P=0.000), VEGF(r=0.50, P=0.000), MMP-2(r=0.50, P=0.000), and MMP-9(r=0.66,P=0.000). DCE-MRI score in malignant breast tumors was statistically significant different higher than benign tumors(P=0.000). As for5points on DCE-MRI score, the sensitivity, specificity, and accuracy were86.5%,94.3%, and88.0%, respectively. The correlation between DCE-MRI score and pathology was significant (r=0.70, P=0.000). The correlation between DCE-MRI score and MVD was significant0=0.52, P=0.000), while VEGF(r=0.44, P=0.000), MMP-2(r=0.42, P=0.000), and MMP-9(r=0.35, P=0.000) were low. Conclusion:3D-CEUS score system with ideal clinical feasibility can comprehensively reflect tumor perfusion and vascular characteristics of benign and malignant breast tumors. Chapter II Evaluation Research on Tumor Response to Neoadjuvant Chemotherapy in Breast Cancer by UltrasonographyPart I Predicting Response to Neoadjuvant Chemotherapy in Breast Cancer by Conventional Ultrasound and Dynamic Contrast-enhanced MRI According to RECISTObjective:To discuss the clinical value in predicting tumor response to neoadjuvant chemotherapy (NAC) in breast cancer by conventional ultrasound (US) and dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI), according to Response Evaluation Criteria In Solid Tumors (RECIST), in order to provide more imaging information for NAC response prediction and treatment scheme formulation. Methods:During the period of May2010through Feb2012, the TNM stage II or III primary breast cancer patients received with neoadjuvant chemotherapy were selected. Prior to NAC, core needle biopsy of the breast cancer was taken under ultrasound guidance. All patients were less than70years old. The primary tumor was evaluated by conventional US and DCE-MRI before and after NAC. They subsequently underwent surgery after completion of4or6cycles of NAC, obtained a satisfied histopathologically evaluation. Esaote Mylab90;(Esaote, Genoa, Italy) was performed, and the measurement of the tumor maximum diameter were carried out by conventional US. DCE-MRI was performed on a dedicated breast magnetic resonance imaging system (Aurora Dedicated Breast MRI Systems, USA). The maximum diameter measurement was carried out on the early phase with highest enhancement by AURORA CAD workstation image post-processing. The tumor response evaluation on conventional US and DCE-MRI according to the RECIST criteria was classified as followed, clinical complete response(cCR), clinical partial response (cPR), clinical stable disease (cSD), and clinical progressive disease (cPD). The clinical response rate(cRR) was calculated as follows, cRR=(cCR+cPR)/the sum of cases×100%. Pathologic response in breast cancer was classified into5grades according to the Miller-Payne (M-P) histopathological grading system. M-P histopathological grading results were divided into two groups, major histological response(MHR)including M-P1, M-P2, M-P3, and Non-major histological response (NMHR), including M-P4, M-P5. The pathologic response rate (pRR) was calculated as follows, pRR=(M-P3+M-P4+M-P5)/the sum of cases×100%. Results:108patients with unilateral breast cancer were included. All were female, aged27to68years, with an average of50.01years old. Of the108patients received NAC, the pathologic response according to the Miller-Payne histopathological grading system was as follows:3cases were Gl,22were G2,41were G3,21were G4, and21were G5. Among108breast cancer patients, according to the RECIST criteria,16cases achieved as cCR,63as cPR,4as cPD, and25as cSD by US assessment. The cRR was73.2%by US. Of21patients with pCR after NAC, the sensitivity, specificity, and accuracy of cCR on US in the diagnosis of pCR were52.4%,94.3%, and86.1%, respectively. The correlation between cCR on US and pCR was significant (r=0.52, P=0.000). Among42cases of MHR, the sensitivity, specificity, and accuracy of cCR on US in the diagnosis of MHR were35.7%,98.5%, and74.1%, respectively. The correlation between cCR on US and MHR was low(r=0.47,P=0.000). For83cases achieved pRR, the sensitivity, specificity, and accuracy of cRR on US in the diagnosis of pRR were81.9%,56.0%, and75.9%, respectively. The correlation between cRR on US and pRR was low(r=0.36, P=0.000).15cases achieved as cCR,71as cPR,2as cPD, and20as cSD by MRI assessment. The cRR was79.6%by MRI. Of21patients with pCR after NAC, the sensitivity, specificity, and accuracy of cCR on MRI in the diagnosis of pCR were47.6%,94.3%, and85.2%, respectively. The correlation between cCR on MRI and pCR was low(r=0.48, P=0.000). The sensitivity, specificity, and accuracy of cCR on MRI in the diagnosis of MHR were31.0%,97.0%, and71.3%, respectively. The correlation between cCR on MRI and MHR was low(r=0.39, P=0.000). The sensitivity, specificity, and accuracy of cRR on MRI in the diagnosis of pRR were86.7%,44.0%, and76.9%, respectively. The correlation between cRR on MRI and pRR was low(r=0.32, P=0.000). Conclusion:The assessment of breast cancer response to neoadjuvant chemotherapy by conventional US and MRI based on RECIST has some limitations in clinical practice. Therefore, a more appropriate and practical standards by imaging modalities in predicting the response to NAC is urgent. Part Ⅱ Comparative Study on Assessment of Breast Cancer Response to Neoadjuvant Chemotherapy by Three-dimensional Contrast Enhanced Ultrasound and Dynamic Contrast-enhanced Magnetic Resonance ImagingObjective:To investigate the changes of the perfusion characteristics and tumor vascular characteristics following neoadjuvant chemotherapy (NAC) by three-dimensional contrast enhanced ultrasound (3D-CEUS), and to discuss the clinical feasibility and diagnostic value of3D-CEUS score in the evaluation of assessment of breast cancer response to NAC, compared with dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI), in order to provide more imaging information for NAC response prediction and treatment scheme formulation. Methods:During the period of May2010through Feb2012, the TNM stage Ⅱ or Ⅲ primary breast cancer patients received with neoadjuvant chemotherapy were selected. Prior to NAC, core needle biopsy of the breast cancer was taken under ultrasound guidance. All patients were less than70years old. The primary tumor was evaluated by3D-CEUS and DCE-MRI before and after NAC. They subsequently underwent surgery after completion of4or6cycles of NAC, obtained a satisfied histopathologically evaluation. Esaote Mylab90(Esaote, Genoa, Italy) was performed with SonoVue (SonoVue, Bracco, Italy), and the measurement of the tumor maximum diameter were carried out by conventional US, according to the RECIST criteria. The changes characteristics in the presence, course and dilated degree of peripheral vessels, the presence of penetrating vessels, the presence and the coarse degree of rim perfusion, and the presence of central vessels, as well as the dilated degree of central vessels, and tumor vascular characteristics following NAC were recorded as Δ3D-CEUS score. DCE-MRI was performed on a dedicated breast magnetic resonance imaging system (Aurora Dedicated Breast MRI Systems, USA) with Gd-DTPA (Magnevist, Germany). The maximum diameter measurement was carried out on the early phase with highest enhancement by AURORA CAD workstation image post-processing, according to the RECIST criteria. Based on the Fischer score, the changes characteristics of DCE-MRI score system focused on enhancement type, time-signal intensity curve (TIC), early phase enhancement rate (EPER), peripheral vessels, and penetrating vessels following NAC were recorded as ΔDCE-MRI score. The alteration of microvessel density (MVD) and vascular endothelial growth factor(VEGF)following NAC were noted as ΔMVD and ΔVEGF score. Pathologic response in breast cancer was classified according to the Miller-Payne (M-P) histopathological grading system. Results:48patients with unilateral breast cancer were included. All were female, aged28to63years, with an average of47.08years old. Of the48patients received NAC, the pathologic response according to the Miller-Payne histopathological grading system was as follows:28cases were G3,12were G4, and8were G5.28patients were NMHR, while20were MHR. The pathologic response rate (pRR) was100.0%. Among48breast cancer patients, according to the RECIST criteria, the sensitivity, specificity, and accuracy of cCR on conventional US in the diagnosis of pCR were62.5%,82.5%, and79.2%, while the sensitivity, specificity, and accuracy of cCR on MRI were55.0%,96.4%, and79.2%, respectively. The sensitivity, specificity, and accuracy of cCR on conventional US in the diagnosis of MHR were75.0%,95.0%, and91.7%, while sensitivity, specificity, and accuracy of cCR on on MRI in the diagnosis of MHR were40.0%,100.0%, and75.0%, respectively. The3D-CEUS score (P=0.000) and DCE-MRI score (P=0.000) decreased statistically significant different after NAC. The sensitivity, specificity, and accuracy of Δ3D-CEUS as for6points in the diagnosis of pCR were100.0%,85.0%, and87.5%,respectively. The sensitivity, specificity, and accuracy of Δ3D-CEUS as for4points in the diagnosis of MHR were100.0%,87.5%, and91.4%, respectively. The sensitivity, specificity, and accuracy of ΔDCE-MRI as for6points in the diagnosis of pCR were100.0%,95.0%, and95.8%, respectively. The sensitivity, specificity, and accuracy of ΔDCE-MRI as for4points in the diagnosis of MHR were100.0%,75.0%, and85.4%, respectively. The correlation between Δ3D-CEUS and ΔMVD was significant (r=0.61, P=0.000), while Δ3D-CEUS and ΔVEGF was low(r=0.59, P=0.000). The correlation between Δ DCE-MRI and ΔMVD was low (r=0.37, P=0.018), while ΔDCE-MRI and ΔVEGF was low(r=0.39, P=0.013). Conclusion:3D-CEUS score system reflected the changes tumor perfusion and vascular characteristics following NAC provides useful imaging information for NAC response prediction and treatment scheme formulation. |
PDF Full Text Request